Compilation and communication strategies for out-of-core programs on distributed memory machines

Rajesh Bordawekar; Alok Choudhary; J. Ramanujam

doi:10.1006/jpdc.1996.0148

Compilation and communication strategies for out-of-core programs on distributed memory machines

Rajesh Bordawekar^*, Alok Choudhary, J. Ramanujam

^*Corresponding author for this work

Electrical and Computer Engineering

Research output: Contribution to journal › Article › peer-review

3 Scopus citations

Abstract

It is widely acknowledged that improving parallel I/O performance is critical for widespread adoption of high performance computing. In this paper, we show that communication in out-of-core distributed memory problems may require both interprocessor communication and file I/O. Thus, in order to improve I/O performance, it is necessary to minimize the I/O costs associated with a communication step. We present three methods for performing communication in out-of-core distributed memory problems. The first method, called the generalized collective communication method, follows a loosely synchronous model; computation and communication phases are clearly separated, and communication requires permutation of data in files. The second method, called the receiver-driven in-core communication, communicates only the in-core data. The third method, called the owner-driven in-core communication, goes even one step further and tries to identify the potential future use of data (by the recipients) while it is in the senders memory. We provide performance results for two out-of-core applications: the two-dimensional FFT code, and the two-dimensional elliptic Jacobi solver.

Original language	English (US)
Pages (from-to)	277-288
Number of pages	12
Journal	Journal of Parallel and Distributed Computing
Volume	38
Issue number	2
DOIs	https://doi.org/10.1006/jpdc.1996.0148
State	Published - Nov 1 1996

Funding

The work of R. Bordawekar and A. Choudhary was supported in part by NSF Young Investigator Award CCR-9357840, by grants from Intel SSD, and in part by the Scalable I/O Initiative Contract DABT63-94-C-0049 from the Advanced Research Projects Agency(ARPA) administered by the U.S. Army at Fort Huachuca. R. Bordawekar is also supported by a Syracuse University Graduate Fellowship. The work of J. Ramanu-jam was supported in part by an NSF Young Investigator Award CCR-9457768, by NSF Grant CCR-9210422, and by the Louisiana Board of Regents through contract LEQSF(1991-94)-RD-A-09. This work was performed in part using the Intel Paragon system operated by Caltech on behalf of the Center for Advanced Computing Research (CACR). Access to this facility was provided by CRPC. Finally, the authors thank the anonymous reviewers for their comments.

ASJC Scopus subject areas

Software
Theoretical Computer Science
Hardware and Architecture
Computer Networks and Communications
Artificial Intelligence

Access to Document

10.1006/jpdc.1996.0148

Cite this

@article{36c2fbae154e4dc8a836d193d32cc011,

title = "Compilation and communication strategies for out-of-core programs on distributed memory machines",

abstract = "It is widely acknowledged that improving parallel I/O performance is critical for widespread adoption of high performance computing. In this paper, we show that communication in out-of-core distributed memory problems may require both interprocessor communication and file I/O. Thus, in order to improve I/O performance, it is necessary to minimize the I/O costs associated with a communication step. We present three methods for performing communication in out-of-core distributed memory problems. The first method, called the generalized collective communication method, follows a loosely synchronous model; computation and communication phases are clearly separated, and communication requires permutation of data in files. The second method, called the receiver-driven in-core communication, communicates only the in-core data. The third method, called the owner-driven in-core communication, goes even one step further and tries to identify the potential future use of data (by the recipients) while it is in the senders memory. We provide performance results for two out-of-core applications: the two-dimensional FFT code, and the two-dimensional elliptic Jacobi solver.",

author = "Rajesh Bordawekar and Alok Choudhary and J. Ramanujam",

note = "Funding Information: The work of R. Bordawekar and A. Choudhary was supported in part by NSF Young Investigator Award CCR-9357840, by grants from Intel SSD, and in part by the Scalable I/O Initiative Contract DABT63-94-C-0049 from the Advanced Research Projects Agency(ARPA) administered by the U.S. Army at Fort Huachuca. R. Bordawekar is also supported by a Syracuse University Graduate Fellowship. The work of J. Ramanu-jam was supported in part by an NSF Young Investigator Award CCR-9457768, by NSF Grant CCR-9210422, and by the Louisiana Board of Regents through contract LEQSF(1991-94)-RD-A-09. This work was performed in part using the Intel Paragon system operated by Caltech on behalf of the Center for Advanced Computing Research (CACR). Access to this facility was provided by CRPC. Finally, the authors thank the anonymous reviewers for their comments.",

year = "1996",

month = nov,

day = "1",

doi = "10.1006/jpdc.1996.0148",

language = "English (US)",

volume = "38",

pages = "277--288",

journal = "Journal of Parallel and Distributed Computing",

issn = "0743-7315",

publisher = "Academic Press Inc.",

number = "2",

}

TY - JOUR

T1 - Compilation and communication strategies for out-of-core programs on distributed memory machines

AU - Bordawekar, Rajesh

AU - Choudhary, Alok

AU - Ramanujam, J.

N1 - Funding Information: The work of R. Bordawekar and A. Choudhary was supported in part by NSF Young Investigator Award CCR-9357840, by grants from Intel SSD, and in part by the Scalable I/O Initiative Contract DABT63-94-C-0049 from the Advanced Research Projects Agency(ARPA) administered by the U.S. Army at Fort Huachuca. R. Bordawekar is also supported by a Syracuse University Graduate Fellowship. The work of J. Ramanu-jam was supported in part by an NSF Young Investigator Award CCR-9457768, by NSF Grant CCR-9210422, and by the Louisiana Board of Regents through contract LEQSF(1991-94)-RD-A-09. This work was performed in part using the Intel Paragon system operated by Caltech on behalf of the Center for Advanced Computing Research (CACR). Access to this facility was provided by CRPC. Finally, the authors thank the anonymous reviewers for their comments.

PY - 1996/11/1

Y1 - 1996/11/1

N2 - It is widely acknowledged that improving parallel I/O performance is critical for widespread adoption of high performance computing. In this paper, we show that communication in out-of-core distributed memory problems may require both interprocessor communication and file I/O. Thus, in order to improve I/O performance, it is necessary to minimize the I/O costs associated with a communication step. We present three methods for performing communication in out-of-core distributed memory problems. The first method, called the generalized collective communication method, follows a loosely synchronous model; computation and communication phases are clearly separated, and communication requires permutation of data in files. The second method, called the receiver-driven in-core communication, communicates only the in-core data. The third method, called the owner-driven in-core communication, goes even one step further and tries to identify the potential future use of data (by the recipients) while it is in the senders memory. We provide performance results for two out-of-core applications: the two-dimensional FFT code, and the two-dimensional elliptic Jacobi solver.

AB - It is widely acknowledged that improving parallel I/O performance is critical for widespread adoption of high performance computing. In this paper, we show that communication in out-of-core distributed memory problems may require both interprocessor communication and file I/O. Thus, in order to improve I/O performance, it is necessary to minimize the I/O costs associated with a communication step. We present three methods for performing communication in out-of-core distributed memory problems. The first method, called the generalized collective communication method, follows a loosely synchronous model; computation and communication phases are clearly separated, and communication requires permutation of data in files. The second method, called the receiver-driven in-core communication, communicates only the in-core data. The third method, called the owner-driven in-core communication, goes even one step further and tries to identify the potential future use of data (by the recipients) while it is in the senders memory. We provide performance results for two out-of-core applications: the two-dimensional FFT code, and the two-dimensional elliptic Jacobi solver.

UR - http://www.scopus.com/inward/record.url?scp=0030295733&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030295733&partnerID=8YFLogxK

U2 - 10.1006/jpdc.1996.0148

DO - 10.1006/jpdc.1996.0148

M3 - Article

AN - SCOPUS:0030295733

SN - 0743-7315

VL - 38

SP - 277

EP - 288

JO - Journal of Parallel and Distributed Computing

JF - Journal of Parallel and Distributed Computing

IS - 2

ER -

Compilation and communication strategies for out-of-core programs on distributed memory machines

Abstract

Funding

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this